The Sunk Cost Fallacy in Trading - When to Take the Loss

Reviewed by Leon Moyer - Published October 9, 2026 - Last reviewed October 9, 2026

The sunk cost fallacy in trading is letting money you have already lost on a position decide what you do with it next. A loss that has already happened is the same whether you close, hold or adjust, so it cannot tell you which of those choices is better. The only fair comparison is what each choice can make and lose from today forward.

Options make the fallacy easy to fall into, because rolling and repairing are always available. How to Roll a Credit Spread and The Stock Repair Strategy each show an adjustment that improves something. This one is about recognizing when an adjustment is really a way to avoid admitting a loss.

Why is a sunk cost irrelevant to your next decision?

Think of the losing trade as two parts:

1. What has already happened. Premium paid or collected, and the change in value since. This is fixed, and it is identical under every choice you could make now.
2. What happens from here. The position you would hold after your decision, with its own maximum profit, maximum loss and break-even.

Only the second part differs between choices, so only the second part can decide between them. In behavioral finance this is the gap between how losses feel and how decisions should be made: closing a trade makes a loss feel real, so traders often prefer any choice that keeps it open.

How does the sunk cost show up in an options roll?

Return to the credit spread from How to Roll a Credit Spread.

This is the test Course 2's When to Roll Options lesson teaches, applied to a losing spread. So the honest question is: with XYZ at $45, would you sell a 45/40 put spread for $1.82, risking $318 to make $182? If yes, the roll is a reasonable trade, and the fact that it follows a loss is beside the point. If no, then rolling is a way to keep a position you would not choose, mainly so the $84 does not have to be called a loss today.

What are the warning signs of a sunk cost trade?

• The plan changed after the loss, not before. If your exit rule said close at a $100 loss and you rolled at $150 instead, the loss made the decision.
• Rolling only to avoid closing. If you would not open the new spread as a fresh trade at today's price, its numbers are not the reason for the roll; the loss is.
• A growing roll count. Each roll looks small, but the trade keeps capital tied up for longer, often at the same or larger risk.
• Tracking only the latest leg. If you judge the trade by whether the newest spread is green, the original loss quietly disappears from your records. Track the whole trade's running total instead.
• "It has to come back." A stock or an option has no memory of your entry price.

How do you decide in advance when to take the loss?

The time to decide how much to lose is before the trade, when nothing is at stake. Course 2's Options Exit Strategy lesson covers writing exits before you enter, and Course 2's How to Track Options Rolls and Adjustments lesson covers keeping one running total across every roll. For a credit spread, the whole-trade maximum loss turns those rules into a number you can check before each roll: the strike width minus the total credit, times 100. In the worked example, rolling out at the same strikes would cut it from $433 to $402, while rolling out and down for a $0.48 debit would raise it to $481. If a roll would push that number past the loss you wrote down at entry, the plan says close. This does not bring the sunk cost back in: the limit is a risk budget set in advance, not a reaction to the loss. A cap on the number of rolls, set before the first one, does the same job for time.

Is taking a loss ever the wrong choice?

Yes, sometimes. If the adjusted position genuinely passes the fresh-trade question, holding or rolling it is a normal trade, not a fallacy. The opposite error also exists: closing a sound position out of fear because it shows a paper loss. The test cuts both ways, because it ignores the past entirely.